Connecting the dots: Relating the infant brain network to infant behavior

In this dissertation, we aimed to explore the relationship between the development of networks in the infant's brain and infant behaviour. Ultimately, asking the question: whether differences in characteristics of infant brain networks could explain differences in social competency and behavioural control. This dissertation specifically focused on social competence and self-regulation during infancy, since both types of behaviour develop considerably during the first year of life. Before we could look into this relationship, we needed to make sure that infant EEG networks were reliable. Therefore, in chapter 2, we looked into the reliability of graph theoretical characteristics of infant EEG networks. We found that global metrics, metrics that are averaged over the entire brain, are generally highly reliable in both the theta and the low alpha frequency bands. Local metrics were less reliable. In chapter 3, we looked into the external factors influencing infant EEG data quality. The factors influencing data attrition described in this study can be broadly divided into three groups: child-related factors, testing-related factors, and longitudinal (study-specific) factors. Three child-related factors were found to influence data loss: gender, age, and head shape. Four testing-related factors were found to influence data loss: time of testing, the season of testing, the research assistant present during the experiment, and task length all had considerable influence on data. Lastly, data attrition rates of the first session of testing were found to be related to the second session of testing, underlining possible longitudinal biases in terms of data loss. After confirming acceptable reliability and data quality of our infant EEG data, we looked into the relationship between infant brain networks and behaviour. In chapter 4, we describe the development of the infant connectome during the first year of life and find a reorganization of the theta network between 5 and 10 months old. After this reorganization, the theta network becomes more responsive towards social cues versus non-social cues. Lastly, in chapter 5, we study whether the infant brain network can predict behaviour and vice versa. We find that infant self-regulation at 5 months old predicts brain network optimization at 10 months old. Conversely, we find that total theta brain network strength at 5 months old predicts self-regulation at 10 months old. Underlining the bidirectional relationship between brain networks and behaviour during development. This dissertation shows the promise of studying infant brain networks to explain infant behaviour. Infant brain network characteristics are reasonably reliable and offer us a unique insight into the optimization of the brain in the first year of life.